报告详情
Deep Learning-Driven Intrusion Detection and Prevention System: A Hybrid Ensemble Framework with Adaptive Thresholding
编号：108
访问权限：仅限参会人
更新：2026-11-10 16:47:25 浏览：7次
Online
报告开始：2026年11月21日 13:00（Asia/Kolkata）
报告时间：15min
所在会场：[S2] Internet of Things and Intelligent Sensing Systems [S2-2] Internet of Things and Intelligent Sensing Systems
摘要
Abstract-This training process lasts until October 2023, giving you a clear image of modern network infrastructures that are increasingly faced with advanced cyber threats, as traditional, signature-based intrusion detection systems (IDS) are not able to defend networks from polymorphic attacks, zero-day attacks and APT-type corporate espionage. Your training data is up to October 2023. We propose a classification framework by mounting an Adaptive Threshold Optimization Module (ATOM) on top of the basic classification models, which dynamically re-alignment all classification thresholds using Bayesian inference and thus lead to a significant drop in false positive rates whilst maintaining high detection sensitivity. Experimental results show that HEDL-IDPS achieves 99.41%, 99.18% and 98.73% detection accuracy. They employ the framework with a false positive rate of 0.31% and mean detection latency of 4.2 ms, enabling real-time deployment in high-throughput enterprise and cloud environments via their framework.
关键词
Intrusion Detection System, Deep Learning, CNN-LSTM Hybrid, Ensemble Learning, Adaptive Thresholding, Cybersecurity, Anomaly Detection, Network Traffic Classification, Bayesian Optimization, Zero-Day Attacks
报告人
Manikandan S.
Professor Trichy； Tamil Nadu；K.Ramakrishnan College of Engineering稿件作者
全部评论
重要日期
-
会议日期
11月20日
2026
至11月22日
2026
-
10月31日 2026
初稿截稿日期
-
11月07日 2026
提前注册日期
-
11月22日 2026
注册截止日期
主办单位
United Societies of Science
承办单位
Pondicherry University
协办单位
IEEE Section
IEEE Madras Section
IEEE Madras Section
联系方式
- ICon INDIA 2026
- ic******@usssociety.org
发表评论